3.6.1 Describe the following causes of errors with reference to an application in each case: data entry, accidental, deliberate, software and hardware. Data entry errors can occur when the data that is supposed to be entered is different from the actual data entered. This can cause problems in the future when looking at that record or performing calculations. One form of data entry is accidental data entry. This can happen if, for example the user that entered the data did not mean to enter the wrong data, such as if he was looking at the wrong paper copy, if he whilst typing he clicked on the button beside the correct one and typed ‘rhe’ instead of ‘the’. Another form of accidental data entry failure can occur if two of the numbers being inserted are switched around, for example ‘725’ instead of ‘752’.
A deliberate cause of error can be the result of malware being installed on the computer by a third party, causing the data inserted to be inaccurate or completely wrong. This can seriously damage computer systems.
A software error can be caused because it is not compatible with the operating system, because the code has not been written properly or because of many other reasons. It can also be that other running applications disallow that software to run properly. A hardware error can occur due to physical misuse such as it falling or being left in a harsh weather condition. This can mean that the data in it can no longer be used because it has been destroyed, or certain parts of that machine no longer work as well as they used to originally. This can also occur due to the natural wear down of certain pieces like the computer cooling fan, etc.
3.6.2 Outline methods of detection and prevention for each of the errors in 3.6.1
Verification : When a system verifies that the data was entered correctly. e.g.
Double entry ( User inputs information twice)- This is used very commonly for Passwords, and sometimes for E-mails.
Proofread ( User reads after (s)he finishes inputting information)
Validation: When a system verifies that the data was entered sensibly. e.g.
Format check ( to check if a name wasn't entered with numbers)
Length Check ( to check a password isn't too short)
Range Checks: Numeric data can be checked in a variety of ways. One way is to apply a reasonable upper and lower bound.
e.g. A date can be checked to ensure that the day is between 1-31 and the months from 1-12.
The typical algorithm for a range check is to use a function that accepts as the data value and values for the upper and lower limits of the range and returns a Boolean true or false depending on the result. A while loop is used to control the process
Data Type Checks: Data values can be checked to see that they match the expected data type.
e.g. a char value for lower case can be checked to see that it falls within the expected integer ASCII values
A variable such as an age can be checked against valid lists of data. For example, a data type of MONTH could contain a list of the valid months. If a month name was entered it could be validated against the contents of the valid list.
Data Format Checks: A string filed for name could be required to conform to the format requirement that at least some letters are input. i.e. the field is required to have a value and not be left blank. Some fields might have a fixed length and some parts of the field may be character. For example, a date format: 14JUL1950. A check could be made that the data is in dd/mm/yyyy format.
Check Digits: Input data or data that is transmitted from one location to another via a network can be validated using the 'check digit' concept.l. Parity checking uses this same concept.
We apply some algorithm the data and from this we determine a number that is appended to the data value. This value is called the 'check digit'. On the entry of the data this value is recalculated and the newly calculated check digit is compared to the original one. If they match, the data is assumed to be correct.
Unweighted Check Digit:
Consider the value 2345
We can find the check digit by adding the digits up: 2+3+4+5= 14.
As 14 is not a single digit we repeat the addition. 1+4=5 - this is our check digit
Thus our data value could be 23455, where the last digit is the check digit.
On input this digit can be recalculated.
If something different is entered (e.g. 22455 the we calculate the check digit 2+2+4+5=13, 1+3=4 ≠ 5 (original check digit) therefore 2245 is not our number.
However this does not pick up on transcription errors e.g. 24355 - 2+4+3+5= 14, 1+4=5, which matches and therefore the error will not be picked up.
Weighted Check Digit:
Weighted check digit works by 'weighing' each digit. We start at the first digit and multiply it by 1, then go to the next digit and multiply it by 2 etc..
e.g.
Say we have the number 2345
Check Digit= 1*5 + 2*4 + 3*3 + 4*2 = 5 + 8 + 9 + 8 = 30 = 3+0 = 3
Therefore our check digit is 3 and the data value would be 23453.
As 4 doesn't match the check digit 3 an error is detected.
3.6.3 Describe methods of recovery from an error.
Backup copies of the data in the hard drive can come in very handy in these situations, although if the error is simple you can simply manually change the appropriate data to solve the problem. Another way of recovering data is through the re-transmission of the original data could also be seen as a possibility.
Best content in StJulians_ComputerScience | Diigo - Groups
Data entry errors can occur when the data that is supposed to be entered is different from the actual data entered. This can cause problems in the future when looking at that record or performing calculations.
One form of data entry is accidental data entry. This can happen if, for example the user that entered the data did not mean to enter the wrong data, such as if he was looking at the wrong paper copy, if he whilst typing he clicked on the button beside the correct one and typed ‘rhe’ instead of ‘the’. Another form of accidental data entry failure can occur if two of the numbers being inserted are switched around, for example ‘725’ instead of ‘752’.
A deliberate cause of error can be the result of malware being installed on the computer by a third party, causing the data inserted to be inaccurate or completely wrong. This can seriously damage computer systems.
A software error can be caused because it is not compatible with the operating system, because the code has not been written properly or because of many other reasons. It can also be that other running applications disallow that software to run properly.
A hardware error can occur due to physical misuse such as it falling or being left in a harsh weather condition. This can mean that the data in it can no longer be used because it has been destroyed, or certain parts of that machine no longer work as well as they used to originally. This can also occur due to the natural wear down of certain pieces like the computer cooling fan, etc.
3.6.2 Outline methods of detection and prevention for each of the errors in 3.6.1
Verification : When a system verifies that the data was entered correctly.
e.g.
Validation: When a system verifies that the data was entered sensibly.
e.g.
e.g. A date can be checked to ensure that the day is between 1-31 and the months from 1-12.
The typical algorithm for a range check is to use a function that accepts as the data value and values for the upper and lower limits of the range and returns a Boolean true or false depending on the result. A while loop is used to control the process
Data Type Checks: Data values can be checked to see that they match the expected data type.
e.g. a char value for lower case can be checked to see that it falls within the expected integer ASCII values
A variable such as an age can be checked against valid lists of data. For example, a data type of MONTH could contain a list of the valid months. If a month name was entered it could be validated against the contents of the valid list.
Data Format Checks: A string filed for name could be required to conform to the format requirement that at least some letters are input. i.e. the field is required to have a value and not be left blank. Some fields might have a fixed length and some parts of the field may be character. For example, a date format: 14JUL1950. A check could be made that the data is in dd/mm/yyyy format.
Check Digits: Input data or data that is transmitted from one location to another via a network can be validated using the 'check digit' concept.l. Parity checking uses this same concept.
We apply some algorithm the data and from this we determine a number that is appended to the data value. This value is called the 'check digit'. On the entry of the data this value is recalculated and the newly calculated check digit is compared to the original one. If they match, the data is assumed to be correct.
Unweighted Check Digit:
However this does not pick up on transcription errors e.g. 24355 - 2+4+3+5= 14, 1+4=5, which matches and therefore the error will not be picked up.
Weighted Check Digit:
Weighted check digit works by 'weighing' each digit. We start at the first digit and multiply it by 1, then go to the next digit and multiply it by 2 etc..
e.g.
Say we have the number 2345
Check Digit= 1*5 + 2*4 + 3*3 + 4*2 = 5 + 8 + 9 + 8 = 30 = 3+0 = 3
Therefore our check digit is 3 and the data value would be 23453.
If we transpose 2 and 3:
new Check Digit= 1*5 + 2*4 + 3*2 + 4*3 = 5 + 8 + 6 + 12 = 31 = 4
As 4 doesn't match the check digit 3 an error is detected.
3.6.3 Describe methods of recovery from an error.
Backup copies of the data in the hard drive can come in very handy in these situations, although if the error is simple you can simply manually change the appropriate data to solve the problem. Another way of recovering data is through the re-transmission of the original data could also be seen as a possibility.
Best content in StJulians_ComputerScience | Diigo - Groups